Analisis Kehandalan Ekstraksi Garis Tepi Bangunan dari Data Foto Udara Menggunakan Pendekatan Deep Learning Berbasis Mask R-CNN

Kristal, Agri; Harintaka, Harintaka

doi:10.12962/j24423998.v17i2.11401

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In conducting the IoU index accuracy test, several parameters such as TP (true positive) and TN (true negative) to represent correct predictions, while FP (false positive) and FN (false negative) are used to represent incorrect predictions [15]. Based on these four components, the precision desired indicators, recall, and F1-score can be calculated for the detection results carried out.…”

Section: Accuracy Test For Segmentation and Classification Modelmentioning

confidence: 99%

Classification Analysis of Industrial Buildings Using Orthophoto and NDSM Data with Deep Learning Approach (Case Study: Kali Rungkut Village, Surabaya)

Raihan,

Rachmadi,

Bimanjaya

et al. 2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

Information of industrial estates is important because it may increase the economy level, industrial goods production, and export activities. In addition, information of industrial areas needs to be identified so that the area does not interfere with agricultural productivity, natural resources, and cultural heritage. To obtain this information, extraction and classification of building footprints using orthophoto data with a deep learning approach is carried out. However, this has the challenge that the condition of the building is highly diverse in both shape and size, so it requires additional data such as height data form (NDSM) to facilitate its identification. The Mask Region-based Convolutional Neural Network (Mask R-CNN) method used for extraction produces 88.49% precision accuracy; 66.82% completeness (recall); and 76.15% F1-score. The classification model performed with the Deep Neural Network (DNN) method, produced excellent accuracy with average values of precision, recall, and F1-score of 0.94; 0.90; and 0.92, respectively.

show abstract

Section: Accuracy Test For Segmentation and Classification Modelmentioning

confidence: 99%

Classification Analysis of Industrial Buildings Using Orthophoto and NDSM Data with Deep Learning Approach (Case Study: Kali Rungkut Village, Surabaya)

Raihan,

Rachmadi,

Bimanjaya

et al. 2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

show abstract

“…The training process of the R-CNN Mask model was carried out based on a training dataset that was run for up to 29 epochs from the trial-error process by enforcing early stopping to avoid overfitting the resulting model [8]. The backbone model used in this study is Resnet-50, as it has a lower error rate for validation and lower complexity [9].…”

Section: Data Processingmentioning

confidence: 99%

Geographic Element Extraction for Supporting Kota Lengkap Using Deep Learning

Kisworini,

Handayani

2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

An indicator of accelerating the ease of investment in land registration as legal certainty and protection of property assets. Pendaftaran Tanah Sistematis Lengkap (PTSL) is an effort to quickly and accurately develop data on all land parcels, both physical and legal. The use of deep learning to speed up mapping is a concrete example of how we are entering a new era of industrial revolution. The advantage of deep learning is to speed up the process of creating a Kota Lengkap map as a manifestation of a complete Pendaftaran Tanah Sistematis Lengkap (PTSL). The limitation in the number of human resources is a significant problem in measuring the road network to realise Kota Lengkap. This can be overcome by using automation to extract the road network. An orthophoto can more accurately show the location of all visible phenomena, thus enabling precise and accurate measurements of area, distance and direction. Deep learning methods can be applied to orthophotos and LIDAR to detect topographic features such as roads or rivers. Using the deep learning method and a split ratio of 80:20, it takes a total of 11 hours 45 minutes to automatically extract road features, while for river features it takes a total of 2 hours 38 minutes, resulting in an overall accuracy of 96.15%. The automated extraction of road and river networks from orthophotos cannot yet be applied directly to planar maps, as it requires post-processing to obtain gapless polygons.

show abstract

Analisis Kehandalan Ekstraksi Garis Tepi Bangunan dari Data Foto Udara Menggunakan Pendekatan Deep Learning Berbasis Mask R-CNN

Cited by 2 publications

References 0 publications

Classification Analysis of Industrial Buildings Using Orthophoto and NDSM Data with Deep Learning Approach (Case Study: Kali Rungkut Village, Surabaya)

Classification Analysis of Industrial Buildings Using Orthophoto and NDSM Data with Deep Learning Approach (Case Study: Kali Rungkut Village, Surabaya)

Geographic Element Extraction for Supporting Kota Lengkap Using Deep Learning

Contact Info

Product

Resources

About